Facebook AI Tool: Facebook is launching a “universal product recognition model”. It uses artificial intelligence. It identifies consumer goods, from furniture to quick vehicles to quick fashion.
It’s the initial move toward a future where the products in every image on its site can be identified and potentially shopped for. “It’s a terrific vision.”
In the near future Product recognition is the first in AI-powered updates going to its e-commerce platforms, says the organization. Eventually, these will combine AI, augmented reality, and even computerized collaborators to create what it calls a “social-first” shopping experience. What’s more, it likewise today launched a feature called Shops, that lets private companies set up free storefronts on Facebook and Instagram.
A key piece of this is Fashion, with the organization suggesting that a future Facebook “AI fashion beautician” could offer users personalized shopping recommendations based on their wardrobe and daily suggestions for the weather to outfits tailored to and their schedule.
Be that as it may, as the Clueless reference suggests, these features aren’t exactly new ideas.
Amazon already manufactured its own AI-powered fashion collaborator with the Echo Look, presently little heard-of. What’s more, utilizing machine vision to identify and search for products has been a reality since at least the Amazon Fire Phone.
Meanwhile, online shopping platforms like eBay already use AI to speed up the process of posting items for sale, and Amazon is one of a number of firms that’s launched its own “Shazam for clothes” utilizing machine learning.
Facebook says what makes its tools different is its scope and exactness. The organization’s new product recognition tool, GrokNet, can identify tens of thousands of different attributes in an image. These range from specific brands to things like shading and size.
Facebook Marketplace has already been deployed GrokNet, users rapidly helped by it, list items for sale by identifying what is in it, and generating short descriptions. You may upload a photograph of your love seat, for example, and Marketplace will suggest posting it as a “black, leather, sectional sofa.”
The organization is additionally testing a version of this tool that’s worked for businesses. When they upload photos to their page containing their own products, the AI system can automatically label them and connect to shopping pages.
In building these tools, Facebook is helped by its access to users’ photos on the Marketplace. GrokNet is trained on a gigantic database on the order of magnitude of around 100 million images, with the larger part taken from Marketplace.
Facebook says this data is fundamental in creating a machine vision system that can identify products in challenging lighting and from dodgy angles — a piece of the online shopping experience that isn’t leaving.
It’s unclear, however, exactly how accurate GrokNet is. The organization says it can identify 90 percent of images on Marketplace in the Home and Garden category, yet it didn’t give comparable statistics for other types of product categories.
As is often the case with tools like this, the difference between the advertised features and real user experience can be huge, and we’ll have to wait and see what reaction GrokNet gets from Facebook’s users.